Thread: Parameter estimation Nelder Mead

So I doubt any of us have a good idea of what all those variables in the matlab code represent in the big picture of the Nelder-Mead method. I can read the C code, but I don't know what it means because I'm not familiar with Nelder-Mead and what the initial simplex should be/do. One thing I can suggest is that you comment out the existing initial_simplex in the C code and write an exact translation of the matlab code. If that makes the numbers match, then you found your discrepancy. I'm assuming you not only can read the matlab code, but know what it means and understand it's purpose in the mathematical sense.

Originally Posted by quzah

but "same old answers" tells me your math is wrong.

Ever given a thought to the idea the matlab answers are wrong?

If you can't put into clear words what you are trying to achieve, then the chances are you can't translate that from one language to the next either.
Quzah.

Yep.

I havent thought about the matlab being wrong, because the papaer I got the original equations from was published, and within the paper it says what a b and c should be - the values my matlab got. I can read the matlab, but I have decided to do some research into the method itself and how it deals with local minimums. As i pointed out in an earlier post the values it is finding are solutions to the problem, just not the solution i require.